EP1523341B1

EP1523341B1 - Fluid purification

Info

Publication number: EP1523341B1
Application number: EP03722811A
Authority: EP
Inventors: John Allen Burrows; John David Yair
Original assignee: Pathogen Solutions (UK) Ltd
Current assignee: PATHOGEN SOLUTIONS (UK) Ltd
Priority date: 2002-05-01
Filing date: 2003-05-01
Publication date: 2012-09-26
Anticipated expiration: 2023-05-01
Also published as: AU2003229971B2; EP1523341A1; CA2483657A1; CA2483657C; US7416588B2; GB0209920D0; AU2003229971A1; WO2003092751A1; US20050173352A1; JP4461204B2; JP2005523810A

Description

The present invention relates to gas purification devices and to methods of gas purification using such devices.
Fluid purification, for example air purification, by means of ultraviolet (UV) light is well known. UV-C (short wave) light having a wavelength of between 100nm and 280nm is generally used in such applications. In particular, germicidal UV-C light having a wavelength of from about 200nm to about 288nm may be used to kill airborne micro-organisms.
A variety of fluid purification, especially air purification, devices are available that attempt to cleanse the fluid of micro-organisms by irradiating the fluid with germicidal UV-C light. However, many such devices pass the fluid over the UV-C light sources in a transverse direction and therefore the length of time during which the fluid is exposed to a significant strength of UV-C light is relatively short. Consequently, the proportion of micro-organisms in the fluid that are exposed to a dose of germicidal UV-C light sufficient to kill is relatively low.
WO00/20045 discloses a device for the treatment of biological fluids. The apparatus has a large diameter passage flow-through UV radiation system with a static mixer system providing an intensive fluid flow mixing within an irradiation area in which the fluid flow is controlled to provide a flow rate not less than a minimum flow rate corresponding to a maximum fluid residence time within said irradiation area required for efficient mixing, and a maximum flow rate providing a minimum residence time for effective inactivation.
US 6322614 discloses a device for high-purity filtering and disinfecting breathing air. The device includes a fine-dust filter and fan arranged on a housing. An elongate air flow tube is also arranged on the housing and is oriented to receive air flow from the housing. A second filter for disinfecting air is included in the elongate air flow tube.
US 5505904 discloses an air disinfection method and unit in which air is passed through a housing from an air inlet to an air outlet opening at a predetermined air flow rate under substantially uniform cross-sectional air flow conditions across closely spaced UV-C irradiation lamps located at a distance of between 2 to 4 inches off centre with the lamps being positioned directly in the air stream so that all air flows past the bulbs at a distance of not more than 2-3 inches.
US 5112370 discloses a device for sterilizing a forced air flow by means of ultraviolet radiation, comprising an elongated housing provided with reflecting inner surfaces accommodating at least an ultraviolet radiation source and a fan for sucking air into the device and sending it out after being subjected to the ultraviolet radiation in an air flow passage.
The problem of increasing the proportion of micro-organisms killed has been attempted to be solved in several devices. For example, the use of UV reflective inner surfaces inside the device has been proposed, in order to increase the level of UV radiation present in the device. Further, elongate UV light sources have been used in purification devices such that the fluid flows along the length of the light source so as to provide a greater exposure time to the UV radiation for each unit of fluid.
However, there remains a need for a fluid purification device that can effectively eliminate from a fluid a very high proportion of micro-organisms such as the anthrax pathogen, bacillus anthracis, which have a lethal UV-C dose of between 10 000 and 15 000 µWscm^-2 in the vegetative form and up to three times this dose when in the spore phase.
The present invention provides a gas purification device, which comprises a chamber having an inlet and an outlet, a fan and four elongate UV-C light sources, wherein said fan causes gas to pass through the chamber from the inlet towards the outlet, the volume of the chamber and the speed of the gas movement caused by the fan being such that the gas has a residence time in the chamber of greater than 1.0 seconds and said light sources being provided in the chamber in an arrangement such that the gas passes along their length and such that each light source forms an elongate edge of a square prism, wherein the UV-C radiation emitted by the UV-C light sources is only of a wavelength of 220nm or above.
Preferably, the four elongate UV-C light sources are of a wattage and a distance from the centre of the chamber such that the level of UV-C energy at any point within the chamber is 15000 µWscm^-2 or more. The UV-C light sources may suitably be 15W or 25W input UV-C light sources.
In one embodiment, the UV-C light sources may be 25W input UV-C light sources and these may be arranged so as to be at a distance from the centre of the chamber of 50mm or less. By selecting appropriate wattage and distances from the centre of the chamber for the light sources and making use of the particular arrangement of four light sources required by the present invention a purification device can be obtained which has a consistently high level of UV-C radiation throughout and which is capable of up to 99.999% reduction of a variety of challenge organisms within 24 hours. The skilled man would be able, without difficulty to select a wattage for the light sources and a size for the chamber such that a particular minimum energy level, for example 15000µWscm^-2, throughout the chamber is achieved
Preferably the elongate UV-C light sources are located close to the inner surface of the chamber. The UV-C light sources are preferably positioned 30 mm or less from the inner wall of the chamber, more preferably 25mm or less, for example 20mm or less. Accordingly, it is preferred that the wattage of the UV-C light sources and the size of the chamber are selected such that the UV-C light sources can be positioned both sufficiently close to the centre of the centre of the chamber so as to provide adequate UV-C radiation throughout the chamber, for example so that the level of UV-C energy at any point within the chamber is 15000 µWscm^-2 or more, and relatively close to the inner wall of the chamber. This allows a device to be provided with improved efficiency for killing micro-organisms.
The four elongate UV-C light sources may be the same or different; preferably they are the same. UV-C light sources other than these four elongate UV-C light sources arranged such that the gas passes along their length and such that each light source forms an elongate edge of a square prism may be included in the device, for example within the chamber. Such additional light sources may be the same or different to the four elongate UV-C light sources. Preferably such additional light sources are the same as the four elongate UV-C light sources.
The UV-C light sources in the device preferably emit germicidal UV-C light in the range of from 240 to 280nm. It is preferred that the UV-C light sources have a peak emission in the range of from 250 to 260nm. More preferably, the UV-C light sources have a radiation peak at about 254nm, for example the UV-C light sources may suitably have a peak emission at 253.7nm.
The UV-C light sources are not vacuum UV sources and the UV-C radiation emitted by the UV-C light sources is only of a wavelength of 220nm or above. The UV-C light sources therefore do not emit in the range of from 160nm to 220nm. UV-C emissions in this region result in the production of monoatomic oxygen. This is undesirable because monoatomic oxygen bonds with O₂ in the atmosphere to produce ozone. Preferably the UV-C light sources do not produce any radiation at a wavelength below 240nm.
The UV-C light sources may suitably be mercury vapour lamps. The UV-C light sources may be any suitable elongate shape, for example they may be elongate with a substantially circular cross section.
The UV-C light sources may suitably be commercially available UV-C light sources such as substantially elongate UV-C lamps. For example, the UV-C light sources may suitably be selected from the Philips TUV range of low pressure mercury lamps, the Philips HTQ range of medium pressure mercury lamps, the Sylvania G-Range of UV-C light sources or a suitable array of light emitting diodes.
The UV-C light sources may be fixed in position such that each forms an elongate edge of a square prism by any suitable means. For example, each light source may be attached to the chamber by means of a bracket to ensure it stays in position.
The fan draws gas through the chamber. It may, for example, be a centrifugal type fan. The fan may suitably be driven by an electric motor. Preferably, the fan is selected so as to have low noise levels, for example the fan may have a noise output of less than or equal to 35 decibels.
The speed of the gas through the chamber, as controlled by the fan, and the volume of the chamber are such that the gas is treated by UV-C radiation for greater than 1.0s, preferably 1.1s or more. To achieve this, the fan moves the gas through the chamber at a slower speed than is conventionally used. Additionally, the arrangement of the UV-C light sources in the device of the present invention allows the gas to be in contact with UV-C radiation for a longer period than conventional devices.
The speed used to achieve the desired UV-C treatment time will of course depend upon the volume of the chamber but may suitably be of the order of from 0.3 to 0.4ms^-1. The skilled man would clearly be able to select suitable combinations of chamber volume and fan speed so as to achieve the desired treatment time without difficulty. It is preferred that the gas moves through the chamber at a rate such that it is treated by UV-C radiation for 1.2s or more, more preferably 1.4s or more, most preferably 1.5s or more, for example 1.7s or more.
It is preferred that the chamber has inner surfaces that are UV-C reflective; preferably, the inner surfaces reflect germicidal UV-C light. More preferably, the inner surface has a coefficient of reflection of 60% or more for germicidal UV-C wavelengths, more preferably 70% or more for example 85% or more. Suitably, the inner surface of the chamber may be aluminium, for example spun aluminium or aluminium alloy. The inner surface of the chamber may be provided with a suitable finish such as a mill finish.
Preferably, the gas purification device includes one or more filters. More preferably, a filter is provided at or near the beginning of the gas flow through the device. For example, a filter may suitably be provided at or immediately adjacent the inlet of the chamber, such that all gas entering the chamber through the inlet subsequently passes through the filter.
Such a filter removes from the gas most or all of the dirt and debris present in the gas to reduce contamination of the chamber and thus reduce the possibility of the luminescence and reflectance of the chamber being unduly affected. In particular, the filter prevents or reduces the amount of dirt and debris settling on the UV-C light sources and thus ensures that the UV-C light sources are kept efficient and operate at design specifications, preferably for the life of the gas purification device.
The filter may be any suitable filter as known in the art. Preferably, the filter has a mesh size such as to catch the majority of dirt and debris present in the gas. However, the filter preferably has a mesh size such that the flow of micro-organisms is not impeded. For example, the filter may have a mesh size of 13.8 pores/cm. The filter may be a removable filter or may be permanently attached to the device. Preferably the filter is removable in order to facilitate cleaning.
Preferably, the gas purification device includes an outer casing within which the chamber is located. It is preferred that the outer casing is made of metal or plastics material.
The device may suitably include means as known in the art for reducing the level of UV light that escapes from the device, for example the device may include baffles.
The device is used to purify a gas such as air.
The present invention also provides a method of reducing the level of microbial contaminants in a gas, which method comprises passing the gas through a device according to the present invention, with the fan of the device causing the gas to move through the device such that the gas is irradiated with germicidal UV-C light from the UV-C light sources.
For the sake of clarity, throughout the specification, the term 'elongate edge of a square prism' refers to an edge of the prism which connects one corner of one square face of the prism with the corresponding corner of the opposite square face of the prism.
An embodiment of the present invention will now be described with reference to the drawings in which:

Figure 1 is a cross section across the width of a device according to the present invention;
Figure 2 is a cross section along the length of the device of Figure 1;
Figure 3 is a graph showing the number of Bacillus globigii spores in a closed system over 8 hours when the system is treated with a device according to the present invention and corresponds with the information shown in Table 1;
Figure 4 is a graph showing the number of Bacillus subtilis spores in a closed system over 8 hours when the system is treated with a device according to the present invention and corresponds with the information shown in Table 2;
Figure 5 is a graph showing the number of Bacillus megaterium spores in a closed system over 8 hours when the system is treated with a device according to the present invention and corresponds with the information shown in Table 3; and
Figure 6 is a graph showing the % reduction in airborne pathogens in a contained room space over 24 hours when the room space is treated with a device according to the present invention and corresponds with the information shown in Table 4.

The gas purification device 1 is a device suitable for purifying air. The device 1 includes a purifying chamber 2 through which air to be purified is passed. The purifying chamber 2 is elongate with a truncated square cross-section, having a length of 516.0mm, a width of 151.4mm, a height of 151.4mm, a cross sectional area of 217cm² and a volume of 11198cm³. The purifying chamber is provided with an air inlet 3 at a first end of the chamber 2a and an air outlet 4 at a second end of the chamber 2b directly opposite the first end, such that in use air flows into the chamber 2 through inlet 3, along the length of the chamber 2 and out of the chamber 2 through outlet 4.
The purifying chamber 2 has an inner surface 5 made from aluminium, which reflects UV-C light. Located within the purifying chamber 2 are four elongate UV-C lamps 6 having 25W input, a radiation peak of 253.7mm and having an emission spectrum such that no radiation is produced below 240nm. Each lamp 6 has a circular cross section of diameter 26.0mm and a length of 416.0mm, with the length of the lamp including end fittings being 450mm.
The lamps 6 run along the length of the chamber 2 from the first end 2a to the second end 2b and therefore in use the air passing through the chamber flows along the length of the lamps 6. The lamps 6 are arranged within the chamber 2 such that each lamp forms one of the four elongate edges of a square prism. The lamps 6 are located so that the surface of each is approximately 50mm from the centre of the chamber 2 and approximately 25mm from the inner surface 5. The lamps 6 are each held in position by means of bracket 9 attached to the inner wall of chamber 2. In operation, the UV-C lamps 6 provide a minimum energy of 15000 µWs cm^-2 throughout the chamber 2.
At the first end 2a of the purifying chamber 2, adjacent the air inlet 3, is provided a fan 7. The fan 7 is powered by an electric motor (not shown) and in use causes air to move through the purifying chamber 2 from first end 2a to second end 2b, with the air entering the chamber 2 via air inlet 3 and exiting via air outlet 4. The fan 7 operates so as to move the air along the length of the chamber 2 at a speed of 0.3ms^-1.
The purifying chamber 2 is provided with filter 10 at the first end 2a of the chamber. The filter 10 has a mesh size of 13.8 pores/cm and is located between air inlet 3 and fan 7 such that all air entering chamber 2 passes through the filter 10. The filter 10 traps airborne particles, thus minimising the amount of airborne dirt and debris entering the chamber 2 and so maintaining the efficiency of the device 1.
The chamber 2 is enclosed within outer casing 8. The outer casing is made of metal and includes means 11 for fastening the device 1 to a surface such as a wall. The airflow within device 1 is managed such that the outer casing 8 does not become hot in use and therefore the exterior of the device 1 can be touched.

EXAMPLES

The device as described above with reference to the drawings was tested in order to demonstrate its ability to purify air.

The organisms employed were as follows unless otherwise specified:

Bacillus megaterium	NCTC 10342
Bacillus globigii	ATCC 49822
Bacillus subtilis	ATCC 19659
Bacillus cereus	NCTC 2599
Salmonella typhi murium	NCTC 74
Ecoli 0157 H7	NCTC 12079 (ATTENUATED STRAIN : EX PUBLIC HEALTH SERVICE CULTURE)
Staphylococcus aureus	NCTC 8532
Aspergilius niger	NCPF 2275

Bacillus megaterium, bacillus globigii and bacillus subtilis are known to be suitable bacillus anthracis surrogates, having similar UV-C susceptibility to that of bacillus anthracis.

Example 1

Volumetric antimicrobial performance

Trials were conducted in a microbiologically sealed PVC construction consisting of two chambers of identical dimension. These chambers were connected horizontally by the device of the present invention as detailed above. However, to permit uniform microbial dispersion and to facilitate study of microbial dynamics in the absence of UV-C doses, the device was modified so that the fan could be run with the lamp out of circuit. Atmosphere was transferred from the first chamber, A to the second chamber, B, by means of the fan incorporated in the device. An atmosphere return tube of 15 cm diameter also connected chambers A and B, giving an overall operating volume of 54m³.
Each chamber contained four floor mounted fans to assist with microbial dispersion and also a silica gel unit to prevent excessive humidity build up. All surfaces (excluding the internal surfaces of the device of the present invention) were sprayed with an anti static treatment. Pressure equalisation occurred via four apertures secured by 0.2 micron membrane filters. The construction was equipped both for the introduction of microbial aerosols (via chamber A) and for volumetric recovery of atmosphere in volumes of an appropriate diluent medium (via chamber B).
Prior to introduction of test organisms the lamp complex was run in circuit for 4 hours to eliminate airborne contamination resident within the system. Control plates showed that on all occasions this conditioning sterilisation action did reduce internal contamination to < 10 cfu/m³.
The test organisms Bacillus globigii, Bacillus megaterium and Bacillus subtilis were obtained as calibrated dry spore suspensions (powder) or as vegetative cultures. Spore suspensions were obtained by heat treatment (63.5°C for 35 minutes) of mid exponential liquid cultures in brain heart infusion containing 1% starch. Heat treated cultures were then lyophilised and assayed. Assays were conducted daily on spore stocks to assure viability (viable titre by enumeration on TSA with confirmation) and vigour (impedance curve; onset of exponential growth and curve slope).
Test organisms in the form of spores were introduced and dispersed in chamber A by a positive air pressure jet while in the case of vegetative cultures dispersion was achieved by use of a fogging device delivering a particle size range of 5-15 micron. The target level of inoculation in all cases was in the 10e7 cfu/m³ range.
All studies were conducted over 8 hours and each study was repeated nine times over consecutive days. All organisms were studied in monoculture. Additionally, as it was predicted that spore precipitation would occur due to gravity or electrostatic attraction, monitoring exercises were conducted with the UV-C lamps out of circuit. Thus background lethality figures due to system artefacts were obtained and this data was employed to correct the lethality data obtained when the UV-C lamps were in operation.
Sampling was achieved by aspiration of a 1m³ atmosphere volume through 100 ml of diluent (peptone saline recovery broth) which formed the initial test dilution. Recovery of isolates was obtained by serial dilution and plating on appropriate agars. All analyses were conducted in duplicate with appropriate controls.
All isolates obtained were confirmed by prescribed biochemical and morphological characteristics.

Results

Tables 1-3 and the related Figures 3-5 summarise the mean data obtained for the 8 hour trial series for each test organism. Each table shows the results for microbial reduction purely due to precipitation or other system artefacts and with UV-C doses corrected for precipitation.

Table 1: Mean data for the recovery of Bacillus globigii in a 54 m ³ closed system over 8 hours with UV-C treatment by the device of the invention.

	Control	UV-C Dosing	Actual % kill
Sample point in hours	Cfu/m³ recovered	Cfu/m³ recovered	% kill corrected for precipitation
0	4.90E+07	5.60E+ 07	0.00
1	4.5E+07	1.50E+07	66.00
2	4.1E+07	3.56E+06	68.40
3	3.8E+07	7.26E+05	72.30
4	3.5E+07	1.22E+05	73.90
5	3.0E+07	1.88E+04	70.40
6	2.6E+07	2.27E+03	74.20
7	2.2E+07	2.68E+02	73.90
8	1.9E+07	3.40E+01	74.50

Overall % kill in 8 hours = 99.9999%
Mean population precipitation rate = 10.8% per hour

Mann-Whitney Test : Bacillus globigii

The median values for Bacillus globigii survival with no UV-C doses and UV-C doses differ significantly
The two-tailed P value is 0.0400 which is considered significant
The P value is exact.

Mann-Whitney U-statistic=17.000
U' = 64.000
Sum of ranks in Column A =109.00. Sum of ranks in Column B=62.000

Parameter:	Column A	Column B
Mean:	2.883E+07	8381041
# of points:	9	9
Std deviation:	1.395E+07	1.851E+07
Std error:	4653075	6171214
Minimum:	4.900	34.000
Maximum	4.55E+07	5.60E+06
Median:	3.020E+07	122000
Lower 95% CI:	1.810E+07	-5849777
Upper 95% CI:	3.956E+07	2.261E+07

Spearman Rank Correlation:

Number of points = 18
Spearman r = -0.6470 (corrected for ties)
95% confidence interval: -0.8594 to -0.2439

r is significantly different than zero. The two-tailed P value is 0.0037, considered very significant.

Table 2: Mean data for the recovery of Bacillus subtilis in a 54 m ³ closed system over 8 hours with UV-C treatment by the device of the invention

	Control	UV-C Dosing	Actual % kill
Sample point in hours	Cfu/m³ recovered	Cfu/m³ recovered	% kill corrected for precipitation
0	5.6E+07	6.1E+07	0.00
1	5.3E+07	1.3E+07	73.30
2	4.9E+07	2.7E+06	72.50
3	4.7E+07	6.4E+05	71.60
4	4.3E+07	1.4E+05	70.10
5	4.0E+07	2.7E+04	71.90
6	3.5E+07	4.5E+03	71.50
7	3.0E+07	5.7E+02	74.10
8	2.6E+07	3.9E+01	78.60

Overall % kill in 8 hours = 99.9999%
Mean population precipitation rate = 9.2% per hour

Mann-Whitney Test : Bacillus subtilis

The median values for Bacillus subtilis survival with no UV-C doses and UV-C doses differ significantly
The two-tailed P value is 0.0400 which is considered significant
The P value is exact.

Mann-Whitney U-statistic=17.000
U' = 64. 000
Sum of ranks in Column A =109.00. Sum of ranks in Column B = 62.000

Parameter:	Column A	Column B
Mean:	3.589E+07	8612457
# of points:	9	9
Std deviation:	1.611E+07	2.009E+ 07
Std error:	5370817	6698000
Minimum:	5.600	39.000
Maximum	5.300E + 07	6.100E+07
Median:	4.000E+07	140000
Lower 95% CI:	2.350E + 07	-6833130
Upper 95% CI:	4.827E + 07	2.40

r is significantly different than zero. The two-tailed P value is 0.0037, considered very significant

Table 3: Mean data for the recovery of Bacillus megaterium-in a 54m ³ closed system over 8 hours with UV-C treatment by the device of the invention.

	Control	UV-C Dosing	Actual % kill
Sample point in hours	Cfu/m³ recovered	Cfu/m³ recovered	% kill corrected for precipitation
0	6.2E+07	7.3E+07	0.00
1	5.8E + 07	2.5E+07	59.30
2	5.4E+07	9.8E+06	54.10
3	5.0E+07	2.3E+06	69.40
4	4.6E + 07	5.0E+05	70.30
5	4.2E+07	1.2E+05	67.70
6	3.7E+ 07	2.2E+04	69.70
7	3.3E+07	2.3E+03	78.10
8	2.6E+07	8.4E+01	74.60

Overall % kill in 8 hours = 99.9999%
Mean population precipitation rate = 10.2% per hour
Mann-Whitney Test : Bacillus megaterium
The median values for Bacillus megaterium survival with no UV-C doses and UV-C doses differ significantly
The two-tailed P value is 0.0400 which is considered significant
The P value is exact.
Mann-Whitney U-statistic = 17.000
U' = 64.000
Sum of ranks in Column A =109. 00. Sum of ranks in Column B = 62.000
Number of points = 18
Spearman r = -0.6470 (corrected for ties)
95% confidence interval: -0.8594 to -0.2439
r is significantly different than zero
The two-tailed P value is 0.0037, considered very significant.
The P value is approximate (exact calculations would have taken too long)
According to the data obtained, and taking into account microbial depletion not due to UV-C doses, greater than 99.999% kill rates were obtained with UV-C doses by the 8 hour mark for all organisms examined. As the initial spore challenge levels were in excess of 1.0 x 10⁷cfu/m³, this represents a 5 log reduction of contaminants over an 8 hour period. Furthermore in the case of each surrogate the microbial reduction obtained by UV-C doses was statistically significantly different from the level of reduction obtained purely by other factors.

Example 2

Performance of the device of the present invention in the reduction of airborne microbial contaminants in a waste handling facility

Trials were conducted employing the space afforded by a laboratory waste handling facility with a working volume of 216m³. The area is employed for the containment and thermal decontamination of Class II Biological waste and has the facility both for the introduction of aerosols and, as detailed above, for the volumetric recovery of atmosphere in sample diluent.
The performance of the device as described above in relation to Figures 1 and 2 in the removal of airborne microbiological contaminants from this space was assessed.
All test organisms were obtained either as calibrated dry spore suspensions (lyophilised powder) or as calibrated mid exponential cultures in brain heart infusion. Spores were introduced and dispersed by a positive air pressure jet while in the case of vegetative cultures dispersion was achieved by use of a fogging device delivering a particle size range of 5-15 microns. As in Example 1 the test environment was conditioned for 4 hours prior to the introduction of test culture.
Throughout the trial the continued uniform atmospheric dispersion of the test organisms was maintained by a series of floor mounted industrial fans. The target level of inoculation in all cases was 10e6/cfu/m³. Dosing trials were conducted to establish loading volumes.
All trials were conducted employing mono cultures and the performance of the unit with each organism was assessed on three separate occasions. The mean data is presented. In this instance no attempt was made to establish loss of culture from the atmosphere due to precipitation. This was because Example 1 had already shown that the lethality of the device was significantly greater in comparison to removal of airborne organisms purely due to precipitation or adhesion effects in the environment.
After charging, the environment was sampled and thereafter on a four hourly basis over a 24 hour period.
The operating target was to obtain a reduction of 99.999% with respect to each target organism.

Table 4 below and the related Figure 6 detail the range of organisms and initial atmospheric loading for each organism employed during the environmental assessment of the device. Additionally, the mean cumulative percentage reduction of atmospheric contamination incremented over the six hour period is given. The table also illustrates the sampling interval at which microbial lethality reached = > 99.999%.

Table 4 Performance of the device in the removal of airborne pathogens in 216 m ³ room space.

	Mean Organism load level cfu/m³	Time in hours; % reduction
Organism		t=0	t=4	t=8	t=12	t=16	t=20	T=24
B.globigii	4.00E+06	0	7.2	26.9	54.6	85.3	98.6	99.999
B.subtilis	3.60E+ 06	0	8.3	21.4	59.3	90.1	97.2	99.999
B.megaterium	7.10E+06	0	7.5	17.5	49.7	83.4	92.0	99.999
S.typhi murium	3.40E+ 06	0	17.4	31.2	91.6	99.999	99.999	99.990
S.aureus	2.90E+06	0	12.2	38.4	82.0	99.999	99.999	99.999
Aspergillus niger	3.10E+06	0	3.7	23.6	73.9	92.7	98.6	99.999
E.coli 0157:H7	5.20E+ 06	0	19.6	31.7	94.3	99.999	99.999	99.999

Within a scope of operation of use over a 24 hour period in an environment with atmospheric contamination commencing at a level in excess of 1.0x10⁶cfu/m³ the dosages of UV-C were sufficient to bring about a 99.999% reduction of all challenge organisms including Bacillus anthracis surrogates by the 24 hour marker.
The data illustrates that the device of the present invention represents a significant advance in atmospheric treatment. The data suggests a sensible and effective combination of airflow rate to UV-C dosage has been conceived in a system which should integrate efficiently and effectively into environmental biohazard protection systems.

Example 3

Performance of the device of the present invention in the reduction of airborne microbial contaminants in a closed system

The performance of the device as described above in relation to Figures 1 and 2 in the removal of airborne microbiological contaminants from this space was assessed.
The following organisms were employed: Staphylococcus aureus; NCTC 11939; carries gentamicin and Chloramphenicol plasmids/epidemic methicillin resistant strain, Staphylococcus aureus; NCTC 11940; epidemic methicillin resistant strain, Staphylococcus_aureus; NCTC 11962; associated with post operative toxic shock.
In vitro trials were conducted by inoculating the surface of Tryptone Soya Agar plates with aliquots of mid exponential cell cultures. All inoculated plates were conditioned at 30°C for 2 hours prior to UV-C treatment.
An exposed agar plate was positioned such that the surface of the plate was 50mm from the UV-C source. Exposure of inoculated plates occurred over successive 15 second increments up to and including the 60 second mark. Cell density was estimated by deployment of sterile bores capable of obtaining 5 cm² sections of agar to a depth of 5 mm. The resulting core was subject to serial dilution with subsequent recovery of isolates on appropriate agars.
Atmospheric volumetric trials were then conducted in a microbiologically sealed PVA construction consisting of a chamber with an operating volume of 54 m^3.. Four floor-mounted fans were employed to assist with microbial dispersion and also a silica gel unit to prevent excessive humidity build up. Pressure equalisation occurred via four apertures secured by 0.2-micron membrane filters. This facility was equipped both for the introduction of microbial aerosols and for volumetric recovery of atmosphere in volumes of an appropriate diluent medium.
All test organisms were obtained as calibrated mid exponential cultures.
Prior to introduction of the test organisms the device was run in circuit for 4 hours to eliminate airborne contamination resident within the system. Control plates showed that all occasions this conditioning sterilisation action did reduce internal contamination to < 10 cfu/m³.
All studies were conducted over 8 hours and each study was repeated twice over consecutive days.
Sampling was achieved by aspiration of a 1m³ atmosphere volume through 100 ml of diluent (Peptone Saline recovery Broth) which formed the initial test dilution. Recovery of isolates was obtained by serial dilution and plating on appropriate agars. All analyses were conducted in duplicate with appropriate controls.

Results

Table 5: Percentage kill (in vitro) verses exposure to UV-C with three strains of Staphylococcus aureus

			Exposure time agar/recovered			cfu/cm2
		T= sec	T= sec	T= sec	T = sec	T = sec	% Reduction
Organism	N	0	15	30	45	60
S.aureus
11939	4	9.30E+ 07	5.20E+ 05	1300	4	0	> 99.999
11940	4	4.60E+07	6.10E+05	2400	9	0	> 99.999
11962	4	1.90E+07	8.30E+ 04	670	3	0	> 99.999

Table 6:UV-C log reduction of three strains of Staphylococcus aureus

	UV-C	UV-C	UV-C	UV-C	UV-C	UV-C
Sample point in hours	cfu/ m3 recovered	Log Kill	cfu/ m3 recovered	Log Kill	cfu/ m3 recovered	Log Kill
	11939	11939	11940	11940	11962	11962
0	3.30E+08	0.0	4.20E+08	0.0	6.10E+08	0.0
1	1.60E+08	0.3	2.70E+08	0.1	3.40E+08	0.0
2	4.50E+07	0.9	6.20E+07	0.7	2.70E+07	1.1
3	1.60E+07	1.3	8.40E+ 06	1.6	1.20E+07	1.4
4	3.20E+06	2.0	9.20E+05	2.6	5.30E+06	1.8
5	9.20E+05	2.6	7.30E+04	3.7	3.00E+05	3.0
6	8.60E+04	3.6	2.00E+04	4.2	2.80E+04	4.1
7	7.40E+02	5.6	3.90E+ 02	5.9	1.70E+02	6.3
8	8.00E+01	6.6	4.00E+01	6.9	2.00E+01	7.2

Treatment by the device of the present invention is clearly shown to be capable of bringing about a greater than 99.999 % reduction of Staphylococcal numbers within one minute. This was achieved with numbers of organisms far in excess of those that would normally be present in a high care medical environment.
The device is also capable of achieving between 6.6 and 7.2 log cycles of kill over an eight hour period. Again this was demonstrated by employing very high numbers of organisms in atmospheric dispersion.

Example 4

Performance of the device of the present invention in the reduction of airborne microbial contaminants in a hospital unit

A field trial was conducted in a four-bed high dependency unit of a London Hospital. The unit volume measured 248m³ and contained four beds and a clinical reception area. The unit was temperature controlled with air handling units fed from external ducts with EU4 primary and EU8 secondary filtering.
The performance of two devices as described above in relation to Figures 1 and 2 was assessed by fitting them in the ward and sampling the air several times per day using Cassala air sampling units. This sampling unit employs the technique of impacting a known volume (200 L) of air onto the surface of a sterile rotating agar plate in a manner that evenly distributes the air borne micro-organisms over the surface of a plate.
The agars employed in this trial were allocated to afford the recovery of a wide range of aerobic airborne bacteria and spores including airborne class II pathogens. A combination of non-selective, elective and selective solid media was employed which included; Typtone soya agar, Violet red bile agar, Violet red bile glucose agar, Brucella medium, Rogosa agar, C.L.E.D agar, MRS agar, Baird Parker agar, DNA-ase agar, and modified forms of these agars.
A control period of 7 days was implemented with the devices switched off (Period A), followed by a test period of a further 7 days with them switched on (Period B). Test plates from the Cassala units were removed, incubated under optimal conditions to afford recovery of visible colonies and examined for evidence of colony forming units. All isolates were grouped and identified according to a scheme involving Gram strain and a series of morphological, biochemical and serological reactions.
In addition the background level of contamination was measured by sampling the air input to the air-handling units on the roof of the building to establish correlation between air input and the effectiveness of the building filter system.

Results

The Mann-Whitney non-parametric T-test has been used as the statistical tool to test significances where applicable for population means.
Data recovered for the mean total viable aerobic count (TVC) in 200 L^-1 of ward atmosphere for each day is presented in Table 7 for both periods A and B. In the same table the respective daily mean counts for total Gram + ve and Gram -ve isolates.

As can be seen, in respect of the Total Viable Count data a highly significant difference exists at the 95 probability level indicating that the average daily TVC cfu/ 200L^-1 of air for period B was lower than period A. In considering the data obtained for Gram +ve and Gram -ve populations a highly significant difference also exists at the 95 probability level that the average counts for both populations is lower during the period (B) when the device was operating.

Table 7: Mean sampling data for categories of organisms isolated from the atmosphere of a high care ward over 2 consecutive 7 day periods with and without the operation of the device

		Internal	Internal	Internal	Internal
		TVC	Gram -ve	Gram + ve	S.aureus
Day	State	cfu 200 L^-1	cfu 200 L^-1	cfu 200 L^-1*	cfu 200 L^-1*
1	OFF	174	93	81	0
2	OFF	231	176	55	0
3	OFF	288	144	144	7
4	OFF	173	87	86	0
5	OFF	324	219	105	3
6	OFF	461	303	158	11
7	OFF	211	113	98	0
1	ON	192	63	129	0
2	ON	78	24	54	0
3	ON	94	28	66	0
4	ON	161	96	65	0
5	ON	67	31	36	0
6	ON	83	48	35	0
7	ON	94	32	62	0
Mean UV-C off		266	162	104	3
			61%	39%	1%
Mean UV-C on		110	46	64	0
			42%	58%	0%

* confirmed bacterial isolates

Taking into account the effect of external intake air microbiological quality, data for Total Viable Count and those obtained for the levels of Gram +ve and Gram -ve contamination are given in Table 8. All categories of count are significantly greater at the 95 % probability level than those obtained in the ward irrespective of whether or not the UV-C device was operating.

Per force the efficiency of the device was measured over two periods (A and B). During Period A (UV-C off) there were lower levels of microbiological input from the external air intake than during period B. However, a significant reduction of microbial loading was still shown in the ward during period B (UV-C on), when the challenge from external air input was greater.

Table 8: Mean sampling data for categories of organisms isolated from the atmosphere of a high care ward External air intake over 2 consecutive 7 day periods showing the TVC data described in Table 7

	intake	intake	intake	intake
Day	TVC	Gram -ve	Gram + ve	S.aureus
	cfu.200litres^-1 *	cfu.200litres^-1 *	cfu.200litres^-1 *	cfu.200litres^-1 *
1	636	438	198	2
2	541	332	209	3
3	506	290	216	0
4	682	386	296	8
5	608	327	281	90
6	930	571	359	114
7	746	459	287	0
8	790	408	382	1
9	870	511	359	3
10	943	633	310	3
11	782	440	342	6
12	907	605	302	0
13	830	430	400	2
14	734	380	354	1
Mean UV-C off	664	400	264	31
		60%	40%	5%
Mean UV-C on	837	396	290	31
		47%	35%	4%

* confirmed bacterial isolates

Included within the observed levels of organisms during Period A was the presence of Staphylococcus aureus. During the second period, Period B with the machine switched on no Staphylococcus aureus was detected.
Over the two periods, it was demonstrated that there was a significant (59%) reduction in the bio-burden even when challenged from a higher external input load.

Claims

A gas purification device (1), which comprises a chamber (2) having an inlet (3) and an outlet (4), a fan (7) and four elongate UV-C light sources (6), wherein said fan causes gas to pass through the chamber from the inlet towards the outlet, the volume of the chamber and the speed of the gas movement caused by the fan being such that the gas has a residence time in the chamber of greater than 1.0 seconds and said light sources being provided in the chamber in an arrangement such that the gas passes along their length and such that each light source forms an elongate edge of a square prism, wherein the UV-C radiation emitted by the UV-C light sources is only of a wavelength of 220nm or above.
A gas purification device according to Claim 1 wherein the four elongate UV-C light sources are of a wattage and a distance from a centre of the chamber such that the level of UV-C energy at any point within the chamber is 15000 µWscm^-2 or more.
A gas purification device according to Claim 1 or Claim 2 wherein the UV-C light sources are 25W input UV-C light sources and are arranged so as to be at a distance from a centre of the chamber of 50mm or less.
A gas purification device according to any one of Claims 1 to 3 wherein the elongate UV-C light sources are positioned 30 mm or less from an inner wall of the chamber.
A gas purification device according any one of Claims 1 to 4 wherein the UV-C light sources emit germicidal UV-C light in the range of from 240 to 280nm.
A gas purification device according to any one of Claims 1 to 5 wherein the UV-C light sources have a peak emission in the range of from 250 to 260nm.
A gas purification device according to Claim 6 wherein the UV-C light sources have a peak emission at 253.7nm.
A gas purification device according to any one of Claims 1 to 7 wherein the UV-C light sources do not produce any radiation at a wavelength below 240nm.
A gas purification device according to any one of Claims 1 to 8 wherein the volume of the chamber is adapted and the fan is adapted such that gas passing through the device is treated by UV-C radiation for 1.2s or more.
A gas purification device according to Claim 9 wherein the volume of the chamber is adapted and the fan is adapted such that gas passing through the device is treated by UV-C radiation for 1.4s or more.
A gas purification device according to Claim 10 wherein the volume of the chamber is adapted and the fan is adapted such that gas passing through the device is treated by UV-C radiation for 1.7s or more.
A gas purification device according to any one of Claims 1 to 11 wherein the chamber has inner surfaces (5) that are UV-C reflective.
A gas purification device according to Claim 12 wherein the inner surfaces of the chamber have a coefficient of reflection of 60% or more for germicidal UV-C wavelengths.
A gas purification device according to Claim 13 wherein the inner surfaces of the chamber have a coefficient of reflection of 70% or more for germicidal UV-C wavelengths.
A gas purification device according to any one of Claims 12 to 14 wherein the inner surface of the chamber is aluminium.
A gas purification device according to any one of Claims 1 to 15 further comprising one or more filters (10).
A gas purification device according to Claim 16 wherein a filter is provided at or near the beginning of the gas flow through the device.
A gas purification device according to Claim 17 wherein a filter is disposed at or immediately adjacent the inlet of the chamber, such that all gas entering the chamber through the inlet subsequently passes through the filter.
A gas purification device according to any one of Claims 16 to 18 wherein the filter has a mesh size of 13.8 pores/cm.
A gas purification device according to any one of Claims 1 to 19 wherein the gas purification device includes an outer casing (8) within which the chamber is located.
A gas purification device according to any one of Claims 1 to 20 wherein the device includes baffles.
A method of reducing the level of microbial contaminants in a gas which method comprises passing the gas through a device according to any one of Claims 1 to 21, with the fan causing the gas to move through the chamber such that the gas is irradiated with germicidal UV-C light from the UV-C light sources.
A method according to Claim 22 wherein the gas is air.